Is it worth using parallel OpenMp sort with an array of 4000 elements? [closed]

Well, it depends. What is the context? If you need to go faster and if you have the parallel resources and if you have the power to run those parallel resources and if you can amortize the costs of development and testing of the code (compared to speeding things up some other way) and if the proposed approach is actually faster on your target hardware then it might be worth it. Only your customers and you can say. If you speed up the sort from 10ms to 3ms I won’t notice if it is part of of a 500ms page render done once a day. But if you execute it 1,000,000 times and I’m waiting for it throughout my working day as central part of my workflow I think I would.

Leave a Comment